U of M Extension Home : U of M Home

Minnesota Crop News > 2001-2008 Archives

For various reasons, crop producers and those who advise them are interested in the amount of phosphate and potash required to raise or increase soil test values for P and K. It would be nice to have standard values that could be used throughout Minnesota.

Unfortunately, changes in soil test values as affect by fertilization are greatly influenced by soil chemical properties. Thinking about phosphorus, some applied phosphorus is fixed or tied up by chemical reactions in the various soils. In acid soils, insoluble iron and aluminum phosphates are formed. Calcium carbonate and related insoluble products are formed when phosphorus is added to calcareous soils (pH of 7.4 or more). In calcareous soils, phosphorus fixation is also influenced by the size of the calcium carbonate particles. Therefore, changes in soil test values for phosphorus can be affected by several soil properties.

Changes in soil test values for potassium are not directly related to soil pH. Instead, the type of clay found in soils has a major impact. Some clay minerals fix (tie up) potassium more readily than others. In addition, soil moisture at the time of sampling can have an impact on soil test potassium. This is especially true if the soil has been dry for an extended period of time before sample collection.

Soil samples are frequently collected after harvest as part of studies where various rates of phosphate and potash have been applied to either the corn or soybean crop. Results from some of these research trials are discussed in the paragraphs that follow.

Information from a study conducted in Wabasha County in 2005 provided an opportunity to measure changes in soil test potassium resulting from potash fertilization. The site was planted to soybeans in 2005. The potash was broadcast and incorporated before planting. Results are summarized in Table 1.

Table 1. Soybean yield and soil test values for K as affected by rate of potash broadcast and incorporated before planting. Wabasha County, 2005.

The soil at this site, a silt loam, was typical of soils throughout southeastern Minnesota. The initial soil test for K measured in a sample collected in the spring was 93 ppm. The soil test results in Table 1 are from samples collected in the fall of 2005.

Broadcast potash produced a small increase in yield. The yields were greater than typical in most years. Comparing the effect of rate of potash on soil test K, the number of pounds of K2O required to change soil test K varied from 8.1 to 13.9 lb. per acre and averaged 10.0 lb. per acre.

A similar trial was conducted in Goodhue County in 2005. Corn was grown on the silt loam soil and potash was broadcast in the spring before planting. Soil samples were collected after corn harvest. The results from this trial are summarized in Table 2. The initial soil test for K was 131 ppm.

Table 2. yield and soil test values for K as affected by rate of applied potash. Goodhue County, 2005.

Calculations of the amount of potash required to raise soil test K by 1 ppm produced different results at this site. The soil texture was a silt loam, very similar to the texture at the Wabash County site. Rates of potash required for the change varied from 6.0 to 8.9 lb. per acre with an average of 7.5 lb. per acre. An exact explanation for the difference between the two sites cannot be identified from the data collected. It is possible that the difference in crops resulting in different amounts of K uptake might be responsible for the difference in requirements.

Clay mineralogy for the soils in central Minnesota is different when soils in southeastern and central Minnesota and compared. In 2005, various rates of phosphate and potash were broadcast and incorporated at two sites before planting soybeans. Soil pH was different. Soil samples collected after harvest were analyzed for P and K. Yields and soil test values for K are summarized in Table 3. The soil pH at the acid site was 6.9 with a soil pH of 7.9 at the calcareous site. The initial soil test for K was 145 ppm at the acid site and 180 ppm at the calcareous site.

Table 3. Soybean yield and soil test values for K as affected by rate of potash broadcast and incorporated before planting.

Considering the initial soil test values for K, a response of soybeans to potash fertilization was not expected and none was measured. The amount of potash needed to bring about a change in soil test K, however, differed with soil pH. For the acid or neutral site, an average of 13.5 lb. K₂O per acre was needed to raise soil test K by 1 ppm. This average value was 7.4 lb. K₂O per acre for the calcareous soil.

Various rates of phosphate were also applied at these sites. The Bray procedure was used to measure soil test P at the acid or neutral site. The Olsen procedure was used to measure soil test P at the calcareous site. These results are summarized in Table 4.

Table 4. Soybean yield and soil test values for P as affected by rate of phosphate broadcast and incorporated before planting.

The initial soil test values were 7 ppm (Bray test) at the acid or neutral site and 10 ppm (Olsen test) at the calcareous site. Considering the results from the samples collected in the fall, the 7 ppm from the Bray test may not have been the true value when samples were collected in the spring. The lack of a yield increase when phosphate was applied also indicates that the initial soil test value for P may have been higher. With a value of 9 or 10 ppm P as measured by the Olsen test, a response to broadcast phosphate was not expected and none was measured.

The amount of phosphate fertilizer required to change soil test P values by 1 ppm were similar. Considering the results from the acid or neutral site, an average of 28 lb. P₂O₅ per acre was needed to change the Bray test by 1 ppm. At the calcareous site, a rate of 27 lb. P₂O₅ per acre was required to change soil test as measured by the Olsen test by 1 ppm.

The soil test P values in Table 4 show changes in soil test P after one growing season. Additional research in Minnesota has provided an opportunity to measure changes in soil test P over a substantial period of time. This research was conducted over a period of 12 years (1974 – 1986). Two rates of P₂O₅ (50 and 100 lb. per acre) were broadcast annually in a corn-soybean rotation. The research was conducted at 3 locations (Crookston, Morris, Waseca).

The amount of phosphate needed to change soil test P by 1 ppm varied with location and rate applied. At Crookston, a change of 1 ppm in the Olsen test required 33 lb. P₂O₅ per acre if the rate was 50 lb. P₂O₅ per acre and 30 lb. P₂O₅ per acre if the rate was 100 lb. P₂O₅ per acre. Results at the Morris and Waseca locations were similar to each other. If the rate was 50 lb. P₂O₅ per acre a rate of 94 lb. P₂O₅ per acre and 103 lb. P₂O₅ per acre at Morris and Waseca respectively was required to increase the Bray P test by 1 ppm. If the P₂O₅ rate was increased to 100 lb. per acre, the amounts changed to 43 and 48 lbs. per acre at Morris and Waseca respectively. These amounts are substantially different from the amounts calculated for when soil test P was measured in the fall following a spring application.

The previous discussion has described how the amount of phosphate required to change soil test P varies with soil pH and the rate of phosphate applied. The rate of change in soil test P, as might be expected, changes with approach used for making fertilizer recommendations. In a 5 year (1999-2004) study conducted at Waseca, 3 approaches to phosphate fertilization (crop removal, U or M broadcast, U of M starter) were compared. This study was conducted in a corn-soybean rotation.

The impact of these strategies on crop yield was summarized in a previous issue of this newsletter. Soil samples (0 to 6 inches) were also collected each year and analyzed for P (Bray procedure). Changes in soil test P over time are shown in Figure 1. The most rapid increase resulted from the crop removal approach to phosphate fertilization. Broadcast recommendations of the U of M produced a gradual increase over time. The soil test P decreased only slightly when U of M guidelines for starter (banded) applications were followed.

The crop removal approach, however, did not increase yield when compared to the U of M broadcast and banded guidelines. Higher rates of phosphate fertilizer were applied when the crop removal approach was used. Since there was no increase in yield, this approach proved to be an expensive way to increase soil test values without any return in added yield.

It would be nice if research could produce guidelines for potash and phosphate use to increase soil test values. However, these changes are affected by many factors in Minnesota and no clear guidelines have been established.

Figure 1. Soil test values for P as affected by strategy used for phosphate application in a corn-soybean rotation at Waseca.

printer-friendly PDF version